dottxt.ai
网页
综合
For inference providers · .txt
…p90, it takes 268 µs on GLM-5.3 and 160 µs on Kimi-K2.6, so every new model needs benchmarking again. .lambda takes 25 and 29 µs. GLM 5.3, batch 32, public benchmark data. We timed each libr…
2026-11-14
自动蜘蛛
相关度 3.7